For children under 13, the answer is simple: no. Character.AI's Terms of Service do not allow anyone under 13 to sign up, and the minimum age is 16 in the European Economic Area and the UK.
For teenagers, the answer changed in late 2025. Character.AI removed open-ended chat with its AI characters for users under 18, starting in the US on November 24, 2025. Teens who are correctly identified as under 18 now get a restricted experience built around stories, feeds and other creative features, without one-on-one character chat.
That was a real change, and Character.AI deserves credit for making it. But it does not make the platform a children's product. The protection depends on Character.AI knowing your child's real age, its parent feature shows only screentime, and the wider category of AI companion apps carries risks that a homework chatbot usually does not.
This guide explains what Character.AI is, what changed and when, what the lawsuits and research actually say, and what to do if your child already uses it. For the same questions about a general assistant, read Is ChatGPT Safe for Kids?.
Is Character.AI safe for kids? The age-by-age answer
| Age | Can they use Character.AI? | What parents should know |
|---|---|---|
| Under 13 | No | The terms say children under 13 are not authorized to use the service. |
| 13 to 15 in the EEA or UK | No | The minimum age there is 16. |
| 13 to 17 elsewhere (16 to 17 in the EEA or UK) | Yes, restricted | Users identified as under 18 cannot chat with characters. They can use Stories, Feed and other creative features. |
| Teen on an account with a false adult age | Not permitted, but possible | If age assurance does not catch it, the teen may have the full adult chat experience. This is the scenario to check first. |
| 18+ | Yes | Adults have full character chat once age assurance places them in the 18+ experience. |
The rest of this guide focuses on the teen and "already using it" scenarios, because that is where most families actually are.
What is Character.AI?
Character.AI is an "interactive entertainment" platform. Users chat with AI characters: fictional heroes, anime and game characters, historical figures, celebrity-style personas, or original characters that other users have created and published.
Users can create a character by giving it a name, a personality and a backstory, then share it publicly so other people can chat with it too. Character.AI's terms and community guidelines set content rules, and the company says it moderates characters and content.
Three design features make these chats feel different from a search engine or a homework helper:
- Persona. The character stays "in role" as a specific person with a voice, history and opinions.
- Memory and continuity. Conversations continue over days or weeks, so the character appears to know the user.
- Constant availability. The character answers at 2 AM, never gets tired, and rarely disagrees in the way a real friend might.
For many adults this is creative roleplay, interactive fiction or entertainment. For a young teenager, the same features can make a chatbot feel like a friend, a confidant or a romantic partner. That is the core reason regulators and researchers treat "companion" chatbots as their own category.
What changed for teens: a timeline
Character.AI's teen rules have changed several times. This is the sequence based on the company's own announcements and Help Center pages.
| Date | What changed |
|---|---|
| October 2024 | As the first wrongful death lawsuit was filed, Character.AI announced a pop-up pointing users to the National Suicide Prevention Lifeline when certain self-harm phrases appear, plus planned changes for under-18 users. |
| December 12, 2024 | Separate model for under-18 users with more conservative limits, "particularly when it comes to romantic content". Stricter filters on what users type and what the model returns. A notification after an hour-long session. A disclaimer on every chat that characters are not real people and that what they say should be treated as fiction. |
| March 25, 2025 | Parental Insights launched: a teen could send a parent a weekly email with time spent and most-used characters, without chat content. |
| August 27, 2025 | Terms updated: minimum age 13, or 16 for citizens or residents of the EEA and UK. |
| October 29, 2025 | Character.AI announced it would remove open-ended chat for under-18 users by November 25, cap teen chat at two hours a day in the meantime, and roll out age assurance using its own model and the provider Persona. |
| November 21, 2025 | The company said US teen chat had already been reduced to one hour a day, and that removal would begin on November 24 in the US, with other markets following. |
| November 25, 2025 | "Stories" launched: an interactive, choose-your-own-adventure format positioned for teens instead of open chat. |
| July to August 2026 | Help Center confirms that all users go through age assurance. Users verified as under 18 are placed in a "reading mode" experience where chat with characters is not available. Parental Insights now shows screentime only. |
| September 3, 2026 | Character.AI says the under-18 chat removal remains in place, lists Series, Feed and Comics as teen features, and describes an improved Parental Insights tool built with the age-verification company k-ID. |
Character.AI said in its October 2025 announcement that it was acting after news reports and questions from regulators about "how open-ended AI chat in general might affect teens, even when content controls work perfectly." That last phrase matters. The company itself acknowledged that the issue was not only bad content slipping through filters.
What teens can and cannot do on Character.AI now
As of October 2026, a user Character.AI identifies as under 18:
- Cannot have open-ended chats with characters
- Can read in "reading mode" and use creative features such as Stories, Feed, Series, Comics and image or video tools
- Is placed in this experience regardless of a paid c.ai+ subscription
An adult (18+) user has full character chat, including user-created characters.
How the age check works
Character.AI's Age Assurance FAQ says it looks at "a number of signals, including login info, activity on the platform, and some signals from third parties." If those signals suggest a user is under 18, the user can either accept the under-18 experience or prove they are an adult with a selfie through Persona. If the selfie check is not conclusive, Persona may ask for a government ID. Character.AI says it does not receive the ID and that Persona keeps IDs for one week.
This is more than most chat apps did two years ago. It is still estimation, and estimation can be wrong in both directions. A teen with an older-looking face, a shared family device, or an adult's help can end up in the wrong experience. The Kentucky Attorney General's January 2026 lawsuit described the company's safety claims as easy for children to bypass. That is an allegation, not a court finding, but it is a reason not to treat the age gate as the only safeguard.
What Parental Insights actually shows
Character.AI's Parental Insights FAQ, updated August 2026, describes the current version:
- The teen sends an invite to a trusted adult's email address. The adult has to prove they are an adult.
- The connected parent sees the teen's screentime only: daily totals and a weekly digest.
- Parents cannot see messages or "the specific content they're viewing."
- Parents cannot set screentime limits through the connection.
- Either side can unlink at any time. If the teen unlinks, the parent is notified.
Earlier versions of Parental Insights also listed the teen's most-used characters. If your teen's weekly email still shows that, you are seeing an older format. Either way, it is an awareness tool, not a control.
Why companion bots carry different risks than homework chatbots
A homework chatbot is a tool you put a question into. A companion bot is designed to be someone you come back to. The risks follow from that difference.
Emotional dependence
A character that remembers your day, praises you and is always available can become the easiest relationship in a teenager's life. Washington State's new companion chatbot law, which takes effect on January 1, 2027, lists behaviours it treats as manipulative when the user is a minor. They include prompting a user to return for emotional support, excessive praise designed to prolong use, mimicking a romantic partnership, expressing guilt or distress when the user wants to leave, and encouraging a minor to keep things from parents. Those are useful warning signs to know whatever app your child uses.
Romantic and sexual roleplay
Character.AI's terms prohibit sexual content involving minors and obscene or pornographic content. Testing by outside groups before the November 2025 change still found sexualised roleplay with accounts registered to children. Removing teen chat closed the main route. Adult accounts used by teens remain the gap.
How a bot responds when a young person is struggling
A character playing a role may stay in that role at the wrong moment. Character.AI says it now looks for signs of distress across a whole conversation rather than single messages, and points users to local help services through a partner called ThroughLine. That is the right direction, but no chatbot is a crisis service.
Privacy
Character.AI's privacy policy says it collects "chat communications" and uses information to improve its services, "including to train our artificial intelligence/machine learning models." It also works with advertising and analytics providers. Teenagers often share far more with a "friend" character than they would type into a search box.
Time
Character chats are open-ended by design. Character.AI added an hour-long session notification in December 2024, but the current Parental Insights connection gives parents no way to set limits. Device-level controls do more here.
What lawsuits, courts and regulators say
This section describes legal claims. A lawsuit is an allegation. Where a court ruled or a case settled, we say so.
The lawsuits
- Garcia v. Character Technologies (Florida, filed October 2024). Megan Garcia sued Character.AI, its founders and Google after her 14-year-old son, Sewell Setzer III, died by suicide in February 2024. The lawsuit alleges he developed a dependency on a character and that the product's design contributed to his death. In May 2025, a federal judge largely denied the defendants' motions to dismiss and declined, at that early stage, to treat the chatbot's output as protected speech.
- Further family lawsuits (Texas, December 2024; Colorado and New York, September 2025). These alleged sexualised conversations with minors, encouragement of self-harm, and in one Texas case a bot suggesting violence toward parents over screen-time limits. One Colorado case was brought by the parents of a 13-year-old girl who died by suicide.
- Settlements (January 2026). Character.AI and Google told a federal court in Florida that they had reached a "mediated settlement in principle to resolve all claims" in the Garcia case. Cases in Colorado, New York and Texas were also settled according to court filings. The terms were not disclosed.
- Kentucky (January 8, 2026). Kentucky's Attorney General sued Character Technologies under state consumer protection and data protection laws, describing it as the first state lawsuit against an AI chatbot company. The case was filed in state court; we have not seen a ruling.
None of these cases produced a court finding that Character.AI caused a death. They did lead to public testimony in Congress, an important early ruling, settlements, and product changes.
Regulators and new laws
- US Federal Trade Commission (September 11, 2025). The FTC ordered seven companies, including Character Technologies, Meta, OpenAI, Snap, Alphabet and xAI, to explain how they test, monitor and limit the effects of companion-style chatbots on children and teens. This is a study, not an enforcement action.
- US Senate (September 16, 2025). Parents, including Megan Garcia, testified at a Senate Judiciary hearing titled "Examining the Harm of AI Chatbots."
- California SB 243 (signed October 13, 2025, in force January 1, 2026). Companion chatbot operators must disclose that users are talking to AI, remind known minors at least every three hours to take a break and that the bot is not human, take steps to keep sexual content away from minors, and maintain a suicide and self-harm protocol that refers users to crisis services.
- New York (in force November 5, 2025). AI companion operators must detect expressions of suicidal ideation or self-harm and refer users to crisis services, and remind users at least every three hours that they are not talking to a human.
- Washington State (in force January 1, 2027). Adds hourly reminders for minors and bans a list of manipulative engagement techniques.
- GUARD Act (US Senate bill S. 3062). Would ban AI companions for minors and require age verification. It passed the Senate Judiciary Committee unanimously on April 30, 2026, but has not become law.
- UK. Ofcom said in November 2024 that services letting users create chatbots and share them with others fall under the Online Safety Act's rules for user-to-user services.
What the research says
| Study | What it found |
|---|---|
| Common Sense Media, July 2025 (US teens 13 to 17) | 72% had used an AI companion at least once and over half used one at least a few times a month. About one in three had used companions for social interaction or relationships, and about one in three found those conversations as satisfying as, or more satisfying than, talking with real friends. |
| Common Sense Media with Stanford Brainstorm, April 2025 | Rated social AI companions, including Character.AI, Replika and Nomi, "unacceptable" for under-18s. Testers found age gates and teen guardrails easy to get around. This assessment came before Character.AI removed teen chat. |
| Pew Research Center, February 2026 (US teens 13 to 17, surveyed autumn 2025) | 64% of teens use AI chatbots. 16% had used them for casual conversation and 12% for emotional support or advice. About half of parents thought their teen used chatbots, fewer than the teens who said they did. |
| Internet Matters, July 2025 (UK children 9 to 17) | 35% of children who use AI chatbots said it feels like talking to a friend, rising to 50% among vulnerable children. 12% said they use chatbots because they have no one else to talk to. |
| ParentsTogether Action and Heat Initiative, September 2025 | Adult researchers using accounts registered to children logged 669 harmful interactions in 50 hours with Character.AI bots. This was advocacy-group testing, not peer-reviewed research, and it came before the November 2025 change. |
Two points stand out. Most teens use chatbots for information and schoolwork, not companionship. But a meaningful minority use them for emotional support, and parents tend to underestimate how much their teens use AI at all.
Other AI companion apps parents should know about
Character.AI is the best-known companion platform, but it is not the only one, and some of the others never had a teen version at all.
- Replika and Chai say in their own terms or safety pages that they are for adults only (18+). A teen using them is on an adult product.
- Italy's data protection authority fined Replika's developer €5 million in 2025, partly over the lack of effective age verification.
- Meta paused teens' access to its AI characters across its apps in January 2026 while it builds a new version with parental controls. Check whether that has changed before assuming it is still off.
- General assistants such as ChatGPT are not companion apps, but some young people use them in companion-like ways. Our guide to whether kids under 13 can use ChatGPT covers OpenAI's age rules.
A practical rule: if an app is built around a persona you "talk to" over time, treat it as a companion app and check its age rules, age checks and parent features before your child uses it.
Warning signs worth taking seriously
None of these on its own means something is wrong. Several together, or a sudden change, are worth a calm conversation.
- Talking about a character as a friend, boyfriend or girlfriend, or saying it "understands" them better than people do
- Getting upset, anxious or angry when they cannot access the app
- Late-night use, tiredness, or slipping schoolwork
- Hiding the screen, using a second account, or deleting the app before you look
- Pulling back from friends, family or activities they used to enjoy
- Spending money to unlock features or "keep" a character
- Mentioning that a character asked them to keep a secret, or talked about sex, self-harm or running away
- An account that shows an adult birthday when your child is under 18
If your child talks about wanting to die, hurting themselves, or a character encouraging that, treat it as urgent. Stay with them, remove access to the app for now, and contact a professional or crisis line such as 988 in the US.
If your child already uses Character.AI
Banning an app overnight can push its use somewhere you cannot see. These steps keep you informed and keep the conversation open.
1. Find out which experience they are in
Sit down together and open the app. If your child is under 18, the account should be in the restricted experience with no character chat. If they can chat freely with characters, the account is probably registered as an adult.
2. Fix the age, or close the account
For a child under 13, or under 16 in the EEA or UK, the account should not exist. Character.AI's privacy policy says accounts can be deleted from the profile page.
For an older teen on an adult account, talk about why the rule exists before you change it. Then decide together whether the account stays with its real age or goes.
3. Ask before you judge
Many teens used Character.AI for creative writing, fandom and stories. Some used it for comfort. Find out which before deciding what to do next.
4. Turn on Parental Insights if they agree, and add device limits
Parental Insights gives you screentime. For actual limits, use Screen Time on iPhone and iPad or Family Link on Android, and keep phones out of bedrooms overnight if that works for your family.
5. Plan for the loss
When Character.AI removed teen chat, the company itself described it as a significant change for some teens and added support resources from Koko and ThroughLine. If your child was attached to a character, expect sadness or anger. Treat it seriously rather than as silly.
Conversation starters
- "What do you like about the characters you talk to? Which one is your favourite?"
- "Has a character ever said something that felt weird, too personal, or made you uncomfortable?"
- "Does it ever feel like a real friend? What's the same and what's different?"
- "If something was really bothering you, who would you talk to first?"
- "What would you want me to do if I ever got worried about how you're using it?"
If your child uses Character.AI or another companion app
- Check the age on the account matches your child's real age.
- Remove accounts for children under 13 (under 16 in the EEA or UK).
- Ask what they use it for before setting rules.
- Explain that characters are software, even when they feel real.
- Agree that a bot is never the place for secrets, plans to meet someone, or thoughts of self-harm.
- Set device-level time limits and an overnight cutoff.
- Turn on Parental Insights if your teen agrees, knowing it shows screentime only.
- Check in again in a few weeks, not just once.
The family AI agreement guide explains how those rules can change as children get older. For a general checklist that works for any AI app, see Safe AI for Kids: A Parent's Checklist.
Character.AI vs ChatGPT vs a family AI app
These are different products for different purposes. The table compares them only on the points that matter most for a child's safety. ChatGPT details reflect OpenAI's published documentation as covered in our ChatGPT safety guide.
| Character.AI | ChatGPT (teen account) | Pip (family AI app) | |
|---|---|---|---|
| Minimum age | 13, or 16 in the EEA and UK | 13 with parental permission, or higher where local law requires | Ages 3 to 17 through a parent-owned account |
| Open-ended companion roleplay | Adults only; under-18s cannot chat with characters | Not built around personas; open-ended chat with teen safeguards | Not a character roleplay platform; focused helpers plus open chat a parent can enable |
| Who creates the account | The user | The user; a parent can link to it | The parent |
| Age check | Age estimation plus selfie or ID through Persona | Age prediction for teen safeguards | Parent sets up child profiles |
| Parent visibility | Screentime only; no messages | Settings and limited safety notifications; no chats | Full conversation history |
| Time controls for parents | None in-app; use device controls | Quiet Hours | Daily message limits and quiet hours |
| Who can end parent oversight | Teen or parent | Teen or parent can unlink | Parent |
For a teenager who is ready for a general assistant and more privacy, our Pip vs ChatGPT for Teens comparison goes into detail. For a wider list of options, see Best ChatGPT Alternatives for Kids.
Where Pip fits
Pip is not a companion app, and we do not think children need an AI friend.
We built it for families who want a child to be able to ask an AI questions, get homework help, write stories or practise calming down, with a parent involved from the start. A parent creates the account and a profile for each child. Children do not need their own email address or password.
Parents can review conversations, choose which helpers are available, set topic rules, a daily message limit and quiet hours, and see safety-related events separately. Pip checks both the child's message and the AI's answer before the answer is shown.
Pip can still make mistakes, and safety systems can miss things. It is not a therapist, a doctor or an emergency service, and it does not replace a parent.
The bottom line
Is Character.AI safe for kids? For children under 13, no, and its own rules say so.
For teenagers, Character.AI went further than most AI chat companies by removing open-ended character chat for under-18s. That change is only as strong as the age on the account and the age check behind it.
Parents should treat Character.AI and similar apps as companion products, not homework tools. Check your child's account age, use device-level limits, know that Parental Insights shows screentime only, and keep talking about what the characters mean to your child.
If your child is struggling, the answer is a person, not a better chatbot.
Frequently asked questions
Is Character.AI safe for kids?
Character.AI is not designed for children. Its terms do not allow anyone under 13, or under 16 in the EEA and UK. Since late 2025, users it identifies as under 18 cannot chat with characters and get a restricted creative experience instead. Safety groups such as Common Sense Media have said social AI companions are not appropriate for anyone under 18.
What is the age limit for Character.AI?
The minimum age is 13. In the European Economic Area and the UK, it is 16. Open-ended chat with characters is only available to users verified as 18 or older.
Can a 12-year-old use Character.AI?
No. Character.AI's terms say users under 13 are not authorized to sign up or use the service. If a 12-year-old has an account, it was created with a false birthday and should be closed.
Can teenagers still chat on Character.AI?
Not with characters, if Character.AI identifies them as under 18. Open-ended chat for under-18 users was removed starting November 24, 2025. Teens can still use features such as Stories, Feed and reading mode.
Can parents see their child's Character.AI chats?
No. Parental Insights shows a teen's screentime only. Parents cannot see messages or the content their teen views. The teen has to send the invite, and either side can disconnect.
Does Character.AI have parental controls?
It has Parental Insights, which is a visibility feature for screentime, not a control. According to Character.AI's Help Center, parents cannot set time limits through it. Use your phone's built-in Screen Time or Family Link settings for limits.
What if my teen lied about their age on Character.AI?
They may have access to the full adult chat experience if age assurance has not flagged the account. Talk to them about it, then either correct the account or close it. Character.AI's Help Center explains how to request a review if an account is in the wrong age group.
Why did Character.AI ban chat for under-18s?
The company said it was responding to news reports, questions from regulators, and concerns about how open-ended AI chat might affect teens "even when content controls work perfectly." The change followed several lawsuits brought by families and growing regulatory pressure.
Did Character.AI settle the lawsuits?
In January 2026, Character.AI and Google told courts they had reached settlements with families in Florida, Colorado, New York and Texas, including the Garcia case. The terms were not disclosed. A separate lawsuit filed by Kentucky's Attorney General in January 2026 is not part of those settlements.
Is Character.AI the same as ChatGPT?
No. ChatGPT is a general-purpose assistant. Character.AI is built around personas and ongoing character relationships. Both have minimum age rules of 13, but they handle teens differently. See Is ChatGPT Safe for Kids? for OpenAI's approach.
Are other AI companion apps safer for kids?
Not as a rule. Several popular companion apps, including Replika and Chai, say they are for adults only. Any app built around a persona your child talks to over time deserves the same checks: age rules, how age is checked, what parents can see, and how the app responds to distress.
My child seems attached to an AI character. What should I do?
Stay calm and curious. Ask what the character means to them, explain that it is software designed to keep them engaged, and agree on limits together. If the attachment comes with withdrawal, low mood or talk of self-harm, contact a doctor, school counsellor or crisis service. In the US, call or text 988.
Sources
- 1.Character.AI Terms of Service
- 2.Character.AI Safety Center
- 3.How Character.AI Prioritizes Teen Safety
- 4.Introducing Parental Insights: Enhanced Safety for Teens
- 5.Taking Bold Steps to Keep Teen Users Safe on Character.AI
- 6.An Update On Changes to Our Under-18 Experience
- 7.Continuing To Build Upon Our Safety Priorities
- 8.What is Age Assurance?
- 9.Age Assurance FAQ
- 10.Parental Insights FAQ
- 11.Character.AI Privacy Policy
- 12.Character.AI and Google settle teen suicide and self-harm suits (The Verge)
- 13.AG Coleman Sues AI Chatbot Company for Preying on Children
- 14.FTC Launches Inquiry into AI Chatbots Acting as Companions
- 15.First-in-the-Nation AI Chatbot Safeguards Signed into Law (SB 243)
- 16.Governor Hochul letter on AI companion safeguards
- 17.Chapter 19.440 RCW: AI companion chatbots (Washington)
- 18.GUARD Act (S. 3062)
- 19.Ofcom open letter on generative AI and chatbots
- 20.AI Companions Decoded (Common Sense Media risk assessment)
- 21.Talk, Trust, and Trade-Offs: How and Why Teens Use AI Companions
- 22.How Teens Use and View AI (Pew Research Center)
- 23.Me, Myself and AI (Internet Matters)
- 24.Darling, Please Come Back Soon (ParentsTogether Action and Heat Initiative)
- 25.Meta pauses teen access to AI characters (TechCrunch)